ABSTRACT

This study examined whether coffee consumption is causally associated with osteoarthritis. A two-sample Mendelian randomization (MR) analysis using inverse-variance weighted (IVW) and weighted median estimates, and the MR-Egger regression method were performed. The publicly available summary statistical datasets of coffee consumption genome-wide association studies (GWASs) meta-analyses on coffee intake from eight Caucasian cohorts (n = 18,176), GWAS meta-analyses of predominately regular-type coffee consumers of European ancestry (n = up to 91,462), and a GWAS in 7410 patients with osteoarthritis in the arcOGEN study with 11,009 controls of European ancestry were evaluated. Four single-nucleotide polymorphisms (SNPs) from GWASs of coffee consumption as instrumental variables (IVs) to improve inference were selected. These SNPs were located at neurocalcin delta (NCALD) (rs16868941), cytochrome p450 oxidoreductase (POR) (rs17685), cytochrome p450 family 1 subfamily A member 1 (CYP1A1) (rs2470893), and neuronal cell adhesion molecule (NRCAM) (rs382140). The IVW method (beta = 0.381, SE = 0.170, p = 0.025) and the weighted median approach (beta = 0.419, SE = 0.206, p = 0.047) showed evidence to support a causal association between coffee consumption and osteoarthritis. MR-Egger regression revealed that directional pleiotropy was unlikely to be biasing the result (intercept = 0.064; p = 0.549), but showed no causal association between coffee consumption and osteoarthritis (beta = − 0.518, SE = 1.270, p = 0.723). Cochran’s Q test and the funnel plot indicated no evidence of heterogeneity between IVestimates based on the individual variants. The results of MR analysis support the observation that coffee consumption is causally associated with an increased risk of osteoarthritis.